Tachykinins belong to a family of short peptides that are widely distributed in the mammalian central and peripheral nervous system (Bertrand and Geppetti, Trends Pharmacol. Sci. 17:255-259 (1996); Lundberg, Can. J. Physiol. Pharmacol. 73:908-914 (1995); Maggi, Gen. Pharmacol. 26:911-944 (1995); Regoli et al., Pharmacol. Rev. 46 (1994)). They share the common C-terminal sequence Phe-Xaa-Gly-Leu-Met-NH2. Tachykinins released from peripheral sensory nerve endings are believed to be involved in neurogenic inflammation. In the spinal cord/central nervous system, tachykinins may play a role in pain transmission/perception and in some autonomic reflexes and behaviors. The three major tachykinins are Substance P (SP), Neurokinin A (NKA) and Neurokinin B (NKB) with preferential affinity for three distinct neurokinin receptor subtypes, termed NK1, NK2, and NK3, respectively. However, functional studies on cloned receptors suggest strong functional cross-interaction between the 3 tachykinins and their corresponding neurokinin receptors (Maggi and Schwartz, Trends Pharmacol. Sci. 18: 351-355 (1997)).
Species differences in structure of NK1 receptors are responsible for species-related potency differences of NK1 antagonists (Maggi, Gen. Pharmacol. 26:911-944 (1995); Regoli et al., Pharmacol. Rev. 46(4):551-599 (1994)). The human NK1 receptor closely resembles the NK1 receptor of guinea-pigs and gerbils but differs markedly from the NK1 receptor of rodents. The development of neurokinin antagonists has led to date to a series of peptide compounds of which might be anticipated that they are metabolically too labile to be employed as pharmaceutically active substances (Longmore J. et al., DN&P 8(1):5-23 (1995)).
The tachykinins are involved in schizophrenia, depression, (stress-related) anxiety states, emesis, inflammatory responses, smooth muscle contraction and pain perception. Neurokinin antagonists are in development for indications such as emesis, anxiety and depression, irritable bowel syndrome (IBS), circadian rhythm disturbances, visceral pain, neurogenic inflammation, asthma, micturition disorders, and nociception. In particular, NK1 antagonists have a high therapeutic potential in emesis and depression and NK2 antagonists have a high therapeutic potential in asthma treatments. NK3 antagonists seem to play a role in the treatment of pain/inflammation (Giardina, G. et al. Exp. Opin. Ther. Patents, 10(6): 939-960 (2000)) and schizophrenia.
Schizophrenia
The NK3 antagonist SR142801 (Sanofi) was recently shown to have antipsychotic activity in schizophrenic patients without affecting negative symptoms (Arvantis, L. ACNP Meeting, December 2001). Activation of NK1 receptors causes anxiety, stressful events evoke elevated substance P (SP) plasma levels and NK1 antagonists are reported to be anxiolytic in several animal models. The NK1 antagonist from Merck, MK-869 shows antidepressant effects in major depression, but data were not conclusive due to a high placebo response rate. Moreover, the NK1 antagonist from Glaxo-Welcome (S)-GR205,171 was shown to enhance dopamine release in the frontal cortex but not in the striatum (Lejeune et al. Soc. Neurosci., November 2001). It is therefore hypothesized that NK3 antagonism in combination with NK1 antagonism would be beneficial against both positive and negative symptoms of schizophrenia.
Anxiety and Depression
Depression is one of the most common affective disorders of modern society with a high and still increasing prevalence, particularly in the younger members of the population. The life time prevalence rates of Major depression (MDD, DSM-IV) is currently estimated to be 10-25% for women and 5-12% for men, whereby in about 25% of patients the life time MDD is recurrent, without full inter-episode recovery and superimposed on dysthymic disorder. There is a high co-morbidity of depression with other mental disorders and, particularly in younger population high association with drug and alcohol abuse. In the view of the fact that depression primarily affects the population between 18-44 years of age e.g. the most productive population, it is obvious that it imposes a high burden on individuals, families and the whole society.
Among all therapeutic possibilities, the therapy with antidepressants is incontestably the most effective. A large number of antidepressants have been developed and introduced to the market in the course of the last 40 years. Nevertheless, none of the current antidepressants fulfill all criteria of an ideal drug (high therapeutic and prophylactic efficacy, rapid onset of action, completely satisfactory short- and long-term safety, simple and favourable pharmacokinetics) or is without side effects which in one or the other way limits their use in all groups and subgroups of depressed patients.
Since no treatment of the cause of depression exists at present, nor appears imminent, and no antidepressant is effective in more than 60-70% of patients; the development of a new antidepressant which may circumvent any of the disadvantages of the available drugs is justified.
Several findings indicate involvement of SP in stress-related anxiety states. Central injection of SP induces a cardiovascular response resembling the classical “fight or flight” reaction characterised physiologically by vascular dilatation in skeletal muscles and decrease of mesenteric and renal blood flow. This cardiovascular reaction is accompanied by a behavioural response observed in rodents after noxious stimuli or stress (Culman and Unger, Can. J. Physiol. Pharmacol. 73:885-891 (1995)). In mice, centrally administered NK1 agonists and antagonists are anxiogenic and anxiolytic, respectively (Teixeira et al., Eur. Pharmacol. 311:7-14 (1996)). The ability of NK1 antagonists to inhibit thumping induced by SP (or by electric shock; Ballard et al., Trends Pharmacol. Sci. 17:255-259 (2001)) might correspond to this antidepressant/anxiolytic activity, since in gerbils thumping plays a role as an alerting or warning signal to conspecifics.
The NK1 receptor is widely distributed throughout the limbic system and fear-processing pathways of the brain, including the amygdala, hippocampus, septum, hypothalamus, and periaqueductal grey. Additionally, substance P is released centrally in response to traumatic or noxious stimuli and substance P-associated neuro-transmission may contribute to or be involved in anxiety, fear, and the emotional disturbances that accompany affective disorders such as depression and anxiety. In support of this view, changes in substance P content in discrete brain regions can be observed in response to stressful stimuli (Brodin et al., Neuropeptides 26:253-260 (1994)).
Central injection of substance P mimetics (agonists) induces a range of defensive behavioural and cardiovascular alterations including conditioned place aversion (Elliott, Exp. Brain. Res. 73:354-356 (1988)), potentiated acoustic startle response (Krase et al., Behav. Brain. Res. 63:81-88 (1994)), distress vocalisations, escape behaviour (Kramer et al., Science 281:1640-1645 (1998)) and anxiety on the elevated plus maze (Aguiar and Brandao, Physiol. Behav. 60:1183-1186 (1996)). These compounds did not modify motor performance and co-ordination on the rotarod apparatus or ambulation in an activity cage. Down-regulation of substance P biosynthesis occurs in response to the administration of known anxiolytic and antidepressant drugs (Brodin et al., Neuropeptides 26:253-260 (1994); Shirayama et al., Brain. Res. 739:70-78 (1996)). Similarly, a centrally administered NK1 agonist-induced vocalisation response in guinea-pigs can be antagonised by antidepressants such as imipramine and fluoxetine as well as L-733,060, an NK1 antagonist. These studies provide evidence suggesting that blockade of central NK1 receptors may inhibit psychological stress in a manner resembling antidepressants and anxiolytics (Rupniak and Kramer, Trends Pharmacol. Sci. 20:1-12 (1999)), but without the side effects of present medications.
Emesis
Nausea and vomiting are among the most distressing side effects of cancer chemotherapy. These reduce the quality of life and may cause patients to delay or refuse, potentially curative drugs (Kris et al., J. Clin. Oncol., 3:1379-1384 (1985)). The incidence, intensity and pattern of emesis is determined by different factors, such as the chemotherapeutic agent, dosage and route of administration. Typically, early or acute emesis starts within the first 4 h after chemotherapy administration, reaching a peak between 4 h and 10 h, and decreases by 12 to 24 h. Delayed emesis (developing after 24 h and continuing until 3-5 days post chemotherapy) is observed with most ‘high-emetogenic’ chemotherapeutic drugs (level 4 and 5 according to Hesketh et al., J. Clin. Oncol. 15:103 (1997)). In humans, these ‘high-emetogenic’ anti-cancer treatments, including cis-platinum, induce acute emesis in >98% and delayed emesis in 60-90% of cancer patients.
Animal models of chemotherapy such as cisplatin-induced emesis in ferrets (Rudd and Naylor, Neuropharmacology 33:1607-1608 (1994); Naylor and Rudd, Cancer. Surv. 21:117-135 (1996)) have successfully predicted the clinical efficacy of the 5-HT3 receptor antagonists. Although this discovery led to a successful therapy for the treatment of chemotherapy- and radiation-induced sickness in cancer patients, 5-HT3 antagonists such as ondansetron and granisetron (either or not associated with dexamethasone) are effective in the control of the acute emetic phase (the first 24 h) but can only reduce the development of delayed emesis (>24 h) with poor efficacy (De Mulder et al., Annuals of Internal Medicine 113:834-840 (1990); Roila, Oncology 50:163-167 (1993)). Despite these currently most effective treatments for the prevention of both acute and delayed emesis, still 50% of patients suffer from delayed vomiting and/or nausea (Antiemetic Subcommittee, Annals Oncol. 9:811-819 (1998)).
In contrast to 5-HT3 antagonists, NK1 antagonists such as CP-99,994 (Piedimonte et al., L. Pharmacol. Exp. Ther. 266:270-273 (1993)) and aprepitant (also known as MK-869 or L-754,030; Kramer et al., Science 281:1640-1645 (1998); Rupniak and Kramer, Trends Pharmacol. Sci. 20:1-12 (1999)) have now been shown to inhibit not only the acute but also the delayed phase of cisplatin-induced emesis in animals (Rudd et al., Br. J. Pharmacol. 119:931-936 (1996); Tattersall et al., Neuropharmacology 39:652-663 (2000)). NK1 antagonists have also been demonstrated to reduce ‘delayed’ emesis in man in the absence of concomitant therapy (Cocquyt et al., Eur. J. Cancer 37:835-842 (2001); Navari et al., N Engl. L. Med. 340:190-195 (1999)). When administered together with dexamethasone and 5-HT3 antagonists, moreover, NK1 antagonists (such as MK-869 and CJ-11,974, also known as Ezlopitant) have been shown to produce additional effects in the prevention of acute emesis (Campos et al., J. Clin. Oncol. 19:1759-1767 (2001); Hesketh et al., Clin. Oncol. 17:338-343 (1999)).
Central neurokinin NK1 receptors play a major role in the regulation of emesis. NK1 antagonists are active against a wide variety of emetic stimuli (Watson et al., Br. J. Pharmacol. 115:84-94 (1995); Tattersall et al., Neuropharmacol. 35:1121-1129 (1996); Megens et al., J. Pharmacol. Exp. Ther. 302:696-709 (2002)). The compounds are suggested to act by blocking central NK1-receptors in the nucleus tractus solitarius. Apart from NK1 antagonism, CNS penetration is thus a prerequisite for the antiemetic activity of these compounds. Loperamide-induced emesis in ferrets can be used as a fast and reliable screening model for the antiemetic activity of NK1 antagonists. Further evaluation of their therapeutic value in the treatment of both the acute and the delayed phases of cisplatin-induced emesis has been demonstrated in the established ferret model (Rudd et al., Br. J. Pharmacol. 119:931-936 (1994)). This model studies both ‘acute’ and ‘delayed’ emesis after cisplatin and has been validated in terms of its sensitivity to 5-HT3 receptor antagonists, glucocorticoids (Sam et al., Eur. J. Pharmacol. 417:231-237 (2001)) and other pharmacological challenges. It is unlikely that any future anti-emetic would find clinical acceptance unless successfully treating both the ‘acute’ and ‘delayed’ phases of emesis.
Viceral Pain and Irritable Bowel Syndrome (IBS)
Visceral sensation refers to all sensory information that originates in the viscera (heart, lungs, GI tract, hepatobiliary tract and urogenital tract), and is transmitted to the central nervous system resulting in conscious perception. Both the vagal nerve via the nodose ganglion and the primary sympathetic afferent nerves via dorsal root ganglias (DRG) and second order neurons in the dorsal horn serve as the initial pathways along which visceral sensory information is conveyed to the brain stem and to the viscero-somatic cortex. Visceral pain may be caused by neoplastic processes (e.g. pancreas cancer), inflammation (e.g. cholecystitis, peritonitis), ischemia and mechanical obstruction (e.g. urether stone).
The mainstay of medical treatment for visceral pain linked to organic disorders (in casu cancer of the viscera) still focuses on opiates.
Recent evidence suggests that non-organic visceral disorders such as irritable bowel syndrome (IBS), non-cardiac chest pain (NCCP) and chronic pelvic pain may originate from a state of “visceral hyperalgia”. The latter is defined as a condition in which physiological, non-painful visceral stimuli (e.g. gut distension) lead to conscious perception of pain due to a decreased threshold for pain. Visceral hyperalgesia may reflect a state of a permanent, post-inflammatory resetting of the threshold for membrane depolarization at neuronal synapses within visceral sensory pathways. The initial inflammation may occur at the periphery (e.g. infectuous gastroenteritis) or at the site of visceral sensory information integration (neurogenic inflammation in the dorsal horn). Both SP and calcitonin gene-related peptide (CGRP) have been shown to act as pro-inflammatory neuropeptides in neurogenic inflammation.
Visceral hyperalgesia is currently considered as one of the prime targets for drug development aimed at treating functional bowel diseases, which occur in 15 to 25% of the western population. They constitute an enormous socio-economic problem in terms of medical care costs, prescription costs and absenteism. Current treatment options include anti-spasmodics (IBS and NCCP), promotility agents (e.g. tegasorod in constipation-IBS), laxatives (constipation-IBS), and loperamide (diarrhea-IBS), amongst others. None of these approaches has been shown to be very effective, particularly in treating pain. Low dose tricyclic antidepressants and SSRIs are used to treat visceral hyperalgesia in pain-predominant IBS, but both classes of compounds may have considerable effects on colonic transit. Ongoing research in this field has identified a considerable number of molecular targets that could serve for drug development in visceral hyperalgesia. These include NK receptors, the CGRP receptor, 5-HT3 receptors, glutamate receptors, and the kappa opioid receptor. Ideally, a “visceral analgesic compound” should block heightened sensory transfer from the viscera to the CNS without affecting the normal physiological homeostasis of the GI tract with regards to propulsive motor activity, absorption and secretion, and sensation. There is compelling evidence linking tachykinin to visceral nociceptive signalling. A number of pre-clinical publications on the role of NK1, NK2 and NK3 receptors in visceral pain and visceral hyperalgesia indicate a discrepancy between the implication of NK1, NK2 and NK3 receptors in the different inflammation hypersensitivity rodent models. Recently, Kamp et al., J. Pharmacol. Exp. Ther. 299:105-113 (2001) suggested that a combined neurokinin receptor antagonist could be more active than a selective neurokinin receptor antagonist. Substance P and NK1, NK2 and NK3 receptors are elevated in clinical pain states, including visceral pain states (Lee et al., Gastroenterol. 118: A846 (2000)). Given the recent failures of NK1 receptor antagonists as an analgesic in human pain trials (Goldstein et al., Clin. Pharm. Ther. 67:419-426 (2000)), combinations of antagonists may be necessary to have a significant clinical effect. NK3 receptor antagonists are anti-hyperalgesic (Julia et al., Gastroenterol. 116:1124-1131 (1999); J. Pharmacol. Exp. Ther. 299:105-113 (2001)). Recently, the involvement of NK1 and NK3 receptors but not NK2 receptors at spinal level was demonstrated in visceral hypersensitivity mediated by nociceptive and non-nociceptive afferent inputs (Gaudreau & Ploudre, Neurosci. Lett. 351:59-62 (2003). Combining the NK1-2-3 antagonistic activity could therefore represent an interesting therapeutic target for the development of novel treatments for visceral hyperalgesia.
A reasonable number of pre-clinical publications over the role of NK1 receptors in visceral pain has been published. Using NK1 receptor knockout mice and NK1 antagonists in animal models, different groups have demonstrated the important role played by the NK1 receptor in hyperalgesia and visceral pain. The distribution of NK1 receptors and substance P favours a major role in visceral rather than in somatic pain. Indeed more than 80% of visceral primary afferent contain substance P compared with only 25% skin afferents. NK1 receptors are also involved in gastrointestinal motility (Tonini et al., Gastroenterol. 120:938-945 (2001); Okano et al., J. Pharmacol. Exp. Ther. 298:559-564 (2001)). Because of this dual role in both gastrointestinal motility and in nociception, NK1 antagonists are considered to have potential to ameliorate symptoms in IBS patients.
Urinary Incontinence
Urge urinary incontinence is caused by urinary bladder or detrusor hyperreflexia (“irritable bladder”). This hyperreflexia relates to hyperexcitability of bladder sensory afferent C-fibers projecting to the spinal cord. The origin of C-fiber hyperexcitability is multifactorial but occurs for example after bladder infection and chronic distention of the bladder wall (eg. benign prostate hypertrophy, BPH). Hence, treatment should be aimed at decreasing neuronal hyperexcitability. Intravesical instillation of vanilloids (eg. capsaicin) results in a long-term beneficial effect on detrusor hyperreflexia refractory to conventional treatment with anticholinergic drugs. Analogous to animal studies, the effect of vanilloids is mediated through a neurotoxic effect on sensory nerve terminals. In human bladder, subendothelial sensory nerves contain tachykinins, which drive detrusor hyperexcitability. The NK receptors involved in this effect are peripheral NK2 receptors and to a lesser extent, also NK1 receptors. The latter are claimed to play a role in bladder hyperreflexia at the level of the spinal cord. As a consequence, a centrally acting NK1/peripherally acting NK2 antagonist is preferred for the treatment of detrusor hyperexcitability. Interestingly, activation of NK2 receptors increases aromatase activity in Sertoli cells. NK2 receptor antagonists reduce serum testosterone levels in mice, and this may be of therapeutic importance in BPH.